However, you didn’t mention how MIMO achieves this faster rate and therefore a possible flaw in your test.  MIMO has multiple antennas to send different packets along different paths therefore effectively allowing simultaneous channels to exist at the same frequency – the important point is this is “spectrally efficient” (lots of data per radio frequency bandwidth) and therefore is likely to be a key technology for the next generation of mobile phones.  I found some good publications on the Airgo Networks website (http://www.airgonetworks.com) – the technology white papers and Airgo’s brochure.  As MIMO relies on the environment to ‘scatter’ the signal and produce these different paths, positioning of the WLAN router and the surroundings will affect the peak performance.

I was surprised to see you didn’t get much of a reduction of bandwidth with a “mixed wireless” solution when compared to a “pure MIMO”.  As MIMO requires multiple antennas on transmitter and receiver (i.e. at the WLAN router and client) there should have been a substantial drop in rate.  As part of my experiments, I did similar experiments using the Belkin Pre-N router with both MIMO and 802.11g PCMCIA network cards in an anechoic chamber (no reflections or interference). In pure MIMO, I recorded 33.6Mbits/sec but with the ‘g’ card, only 6.4 Mbits/sec.

I do agree with your warning as to the incompatibilities of these systems and it is a shame that the IEEE haven’t agreed on the full 802.11n standard.  When they do, MIMO systems will come into their own and give us all the wireless bandwidth we currently enjoy with wired systems.

David Wright